IRLE

IRLE WORKING PAPER#13-88

November 1988

Barry Eichengreen

Unemployment In Inter-War Britain

Cite as: Barry Eichengreen. (1988). “Unemployment In Inter-War Britain.” IRLE Working Paper No. 13-88. http://irle.berkeley.edu/workingpapers/13-88.pdf

irle.berkeley.edu/workingpapers

eScholarship provides open access, scholarly publishingservices to the University of California and delivers a dynamicresearch platform to scholars worldwide.

Institute for Research on Labor andEmploymentUC Berkeley

Title:Unemployment in Interwar Britain

Author:Eichengreen, Barry, University of California, Berkeley

Publication Date:11-01-1988

Series:Working Paper Series

Publication Info:Working Paper Series, Institute for Research on Labor and Employment, UC Berkeley

Permalink:http://escholarship.org/uc/item/5848x6z8

Keywords:Eichengreen, Britain, unemployment

Copyright Information:All rights reserved unless otherwise indicated. Contact the author or original publisher for anynecessary permissions. eScholarship is not the copyright owner for deposited works. Learn moreat http://www.escholarship.org/help_copyright.html#reuse

Unemployment in Interwar Britain

Barry Eichengreen

University of California at Berkeley

November 1988

Forthcoming in Recent Findings in Economic and Social Research. The NationalScience Foundation and Earhart Foundation provided support for the researchreported here.

UNEMPLOYMENT IN INTERWAR BRITAIN

It was the autumn of 1929 when business began to decline. Where Neil

Smith's employer had needed six lorry drivers, now he needed only five.

But Smith, a young Englishman, was not worried. With his employer's

cooperation, the six drivers formed a "pool" to share the remaining work.

His foreman arranged to have a different member of the pool "play off" (be

temporarily laid off) for three days at a time. When his turn came

around, Smith signed on at the Employment Exchange. Having been

unemployed for two three-day periods over the last two months, he had

already registered at the Exchange and so qualified immediately for

unemployment benefit. The 13 shillings he and his young wife received for

each three-day period were not much less than his wages while employed,

and even permitted a visit to the cinema during his short "vacation."

It was the winter of 1935. Michael Richards had been without steady

work for nearly five years. Initially confident of his ability to find

another job, after months of frustration he grew sullen and despondent.

At first, his 32 shillings a week in benefit had been enough, after rent

and expenses, to put food on the table for his wife and three children.

But after exhausting his entitlement to unemployment insurance benefit,

Richards was forced to turn to the meagre 22 shillings of means-tested

relief provided by the Public Assistance Committee. Potatoes and turnips

replaced meat on the table, and even that was in short supply. Smith's

children seemed small compared to their classmates. He himself blamed

inadequate diet for the hacking cough which had caused many a foreman to

turn him away. The search for work being futile, he spent most of the day

on the street commiserating with his mates.

Messrs. Smith and Richards are fictional characters. They are

representative of two caricatures of unemployment in interwar Britain.

One portrays interwar unemployment as an economic and social catastrophe

of unparalleled dimensions. Unemployment in this view resulted from the

collapse of the market economy in the post-1929 Slump. Inability to find

work was not the fault of the individual but of the circumstances in which

he found himself. At least this was true until the experience of

unemployment began to exercise its pernicious effects. In addition to its

effect on health, unemployment threatened the psychological well-being of

those who endured it. -The unemployed worker became demoralized, depressed

and undesirable from potential employers' point of view. Consequently

there emerged by the second half of the 1930s a large pool of long-term

unemployed for whom idleness was virtually a permanent condition.

The second view characterizes interwar unemployment as a largely

voluntary practice in which workers were encouraged to engage by Britain's

inordinately generous unemployment insurance system. Often, benefits

amounted to a substantial share of wages and could be drawn without any

effective waiting period. Employers like Neil Smith's were encouraged to

adopt the 0X0 System (so called because the Ministry of Labour used O's to

denote days of employment and X's to denote days unemployed), and persons

out of work were induced to remain unemployed in the hope that a more

desirable position might turn up. From this perspective, unemployment was

not pernicious because workers chose it voluntarily. If it had costs to

society, in the form of the lost output of the voluntarily unemployed,

these were attributable to an ill-devised system of unemployment

insurance.

Much recent research on interwar unemployment has revolved around

these two views. It is easy to see why the topic should have attracted so

much attention and has been the subject of such controversy. The

emergence of high unemployment has been the most disturbing feature of the

British economy in the 1980s, and its persistence has been the most

important failure of Margaret Thatcher's economic policies. To better

understand both the causes and effects of high unemployment, economists

and historians have turned to the only other period in the 20th century

when unemployment has been so widespread and persistent: the 1920s and

1930s. Increasingly, many of the same debates which occupy the pages of

our newspapers also fill the pages of historical journals.

This research by historians has revealed that not one but both

stereotypes of interwar unemployment have considerable validity. At any

moment in time, there were a large number of individuals who moved back

and forth between work and dole, experiencing a succession of short spells

of unemployment. At the same time, there was a hard core of long-term

unemployed, particularly in the 1930s, for whom prospects of re-employment

were dim. In effect, the British labor market was divided into two

segments, one in which turnover was rapid and movements into and out of

the pool of unemployed occurred smoothly, and a second in which both

employment and unemployment were long term, so that once in either state

an individual had little chance of getting out. How these two situations

could coexist remains the major puzzle for historians of interwar

unemployment.

* * *

The first problem that any study of unemployment must confront is

measurement. Even today there is no generally accepted definition of

3

unemployment. Is a man who works half a day employed or unemployed? What

about a married woman who, upon losing factory employment, spends her day

engaged in housework? Even when one adopts a particular definition,

there is the problem of accurately enumerating those in and out of work.

For interwar Britain, the generally accepted statistics on unemployment

derive from the operation of the unemployment insurance system. The

unemployment rate among the insured is shown in Figure 1. Not all workers

were covered by the system; most notably, agricultural workers, the

self-employed, and new entrants to the labor market who had not yet

qualified for insurance do not appear in the statistics underlying Figure

1. Moreover, insured persons who had been denied or exhausted their

benefits might drop off the Ministry of Labour's register and not be

counted as unemployed. The only other comprehensive source of

information, the 1931 Census, suggests that if the uninsured are included

then unemployment may have been several percentage points lower throughout

the period. But there is no reason to suspect that Figure 1 presents a

particularly misleading picture of fluctuations in unemployment over time.

There is no dearth of explanation for those fluctuations. The most

prominent feature of Figure 1 is the dramatic rise in unemployment after

1929. Here it is argued that the worldwide depression created a crisis

for the British economy. Britain, after all, was an export economy, and

once the depression caused the collapse of British exports, a contraction

of employment opportunities was inevitable. Some interwar economists,

like A.C. Pigou, while acknowledging the severity of the Depression

objected that unemployment was by no means the inevitable result. If

laborers had been willing to work for less, the prices of British exports

could have been reduced, more could have been sold abroad, and employment

dHlNfl

8£6i i£ n ss n ee BE ie ee 62 BE is 9E SE w ES BEST

8'8T

fi'ET

8'fiT

E'AT

8'8E

X

8S6T-SS6T 'SH3HHOH Q3HnSHI HOI 3IW iH3H,1i01dH3Hn 'I 'DII

would not have had to contract. The problem, as they saw it, was that

real wages (money wages deflated by product prices) rose dramatically

after 1929 instead of falling (as shown in Figure 2). This resulted not

from a rise in money wages but because prices collapsed and wages lagged

in following them down. Some critics, like Pigou, blamed the unions for

the slow adjustment of wages. Others emphasized the extent to which wages

were governed by custom rather than current economic conditions. Still

others, like John Maynard Keynes, in his General Theory of Employment,

Interest and Money, argued that wage cuts would only aggravate the

situation, since the problem was not that labor was too costly but that

the demand for products was too low and reducing wages would further

reduce consumption.

A second prominent feature of the unemployment series in Figure 1 was

its persistently high level. Unemployment had reached high levels well

before the interwar depression. Something other than the post-1929 slump

had to be invoked to explain what was by then already a decade-old

problem. Hence the appeal of the argument that unemployment was

subsidized by the generous provision of insurance benefits, a position

argued recently by the economists Daniel Benjamin and Levis Kochin [1].

There is no disputing that, from 1922, unemployment benefits for a family

of four represented a substantial fraction of average adult male wages

(see Table 1). But that fraction was also substantially less than one so

that unemployment did lower incomes. The level of benefits alone cannot

tell us much about the extent to which the dole encouraged unemployment.

Benjamin and Kochin's statistical analysis led them to conclude that over

the period as a whole the operation of the unemployment insurance system

raised unemployment by five to eight percentage points. Their estimate

mm sow im

8E i£ 98 S£ H ££ EE IS 8E 8E 9E S2

8£6T-E86T 'S3M 3D«H A3HOH flH« TOH '2 '

EE

£8

86

8TT

£TT

68

II6

Table 1

Year

1920

1921

1922

1923

1924

1925

1926

1927

1928

1929

1930

1931

1932

1933

1934

1935

1936

1937

1938

Wages,

AverageWeeklyWages

73.8

70.6

59.1

55.5

56.0

56.4 '

55.8

56.2

55.7

55.8

55.7

54.9

54.0

53.7

54.3

55.0

56.1

57.2

58.9

Benefits and UnemploymentUnited Kingdom 1920-38

AverageWeeklyBenefits

11.3

16.8

22.0

22.0

23.7

27.0

27.0

27.0

27.7

28.0

29.5

29.5

27.3

27.3

28.6

30.3

32.0

32.0

32.8

Rates

Benefits/Wages

.15

.24

.37

.40

.42

.48

.48

.48

.50

.50

.53

.54

.50

.51

.53

.55

.57

.56

.56

UnemploymentRate

3.9

17.0

14.3

11.7

10.3

11.3

12.5

9.7

10.8

10.4

16.1

21.3

22.1

19.9

16.7

15.5

13.1

10.8

12.9

Notes: Wages (for adult males) and benefits (for typical family of four)in shillings per week. Unemployment rate in per cent.

has been vigorously attacked by historians. My own research [2] suggests

that the effects of unemployment benefits were considerably smaller.

Factors other than the insurance system must have contributed to the high

average level of unemployment over the period.

Among other factors historians have long emphasized the role of

structural (or "mismatch") unemployment. Before World War I, one in four

British workers had been employed in the staple trades: mining,

shipbuilding, cotton manufacture, mechanical engineering and iron and

steel. During the war, domestic and foreign capacity in all these

industries was considerably expanded. Hence in the 1920s the British

staple trades suffered from intense foreign competition and low prices.

The problem for the economy was to redeploy resources to other sectors.

Between 1920 and 1925 employment in the staple industries fell by over one

million. But the process of redeployment was not smooth. Even a coal

miner fortunate enough ultimately to find a job in the motor industry

might suffer an extended spell of unemployment in the interim. As late as

July 1929 the five staple trades accounted for nearly half the insured

unemployed. N.F.R. Crafts, Mark Thomas and Mary MacKinnon have recently

shown that the average level of unemployment was raised quite considerably

by the extent of structural imbalance.

Part of the difficulty of redeploying workers from declining to

expanding sectors was the concentration of unemployment in certain

depressed regions. In June 1936 unemployment varied from 32 per cent in

Wales, to 17 per cent in Scotland and the North-west, to 5 per cent in the

South-east. One might think that this was simply another symptom of

structural unemployment: since coal mining was depressed and since the

industry was concentrated in Wales, it followed that Wales should have

been depressed. This was not the entire story, however: unemployment

varied across regions even after eliminating the effects of industrial

mix. These regional differentials created further difficulties for the

operation of the labor market. Moving in search of employment was both

costly and risky. As one Londoner put it, "You live on more than just

what you get from your work. There is the fact that you're known in the

neighborhood and can find help tiding you over bad times. Then there is

your family. ..It's pretty risky to move away and leave all that for a job

that just might be a good one."

* * *

Just as unemployment was spread unevenly across industries and

regions, it varied across demographic groups. Typically unemployment

rates were almost twice as high for men as for women, and for males aged

55-59 as for those aged 18-24. Comparing adults of different ages,

unemployment traced out a U-shaped pattern, starting at high levels among

young adults, declining to low levels among workers in their late 30s and

early 40s, and then rising among older workers. The relatively young

complained that employers discriminated against the inexperienced, while

the elderly complained of "the wall of age" — that employers turned away

older workers on the grounds that they had come to expect higher wages,

that they might be more subject to illness, or that they might be

disinclined to learn new methods.

One reason for the lower unemployment rates among women was not that

they really fared better but that upon becoming unemployed they were more

inclined to drop out of the labor force and to disappear from the

statistics. In addition, Carol Heira [5] has pointed to the fact that many

new firms in expanding sectors preferred to take on new workers like women

and juveniles without prior experience in industry. These workers were

desirable by virtue of the fact that they did not bring to employment

entrenched attitudes about work pace and organization.

Youth unemployment rates were lowest of all. Where the overall

unemployment rate among insured persons averaged 14 per cent between 1920

and 1938, juvenile unemployment, for those under eighteen years of age,

averaged only 5 per cent. This contrasts provocatively with the 1980s,

when youth unemployment rates have been much higher, not lower than those

of adults. Why should .the youth/adult unemployment differential have

risen so dramatically over the last 50 years? My work [3] points to the

possibility, as for women, that the low levels of youth unemployment

between the wars were partly a figment of the data, due to the difficulty

of accurately enumerating unemployed juveniles actually searching for

work. But interwar unemployment rates for juveniles were exceptionally

low for a number of additional reasons, including demographic factors (a

lower ratio of juveniles to adults in the population than in the 1980s),

the industrial composition of employment (which favored industries

employing a large share of juveniles in their work forces), and a lesser

reliance on seniority as the basis for layoffs (so-called "inverse

seniority" layoff rules).

A remarkable feature of this labor market was the amount of movement

in and out of unemployment. Mark Thomas [6] has estimated that in the

early '30s one out of every 15 unemployed persons found a job in a given

week. Workers moved in and out of the pool of unemployed at two to three

times the rates of the 1980s. This meant that the burden of unemployment

was widely shared. In a year like 1932, when the unemployment rate was 22

8

per cent, not 22 but 53 per cent of workers experienced at least some

unemployment.

Part of the explanation for the speed of turnover lies in the

operation of the unemployment insurance system, which gave rise to

arrangements like the 0X0 system described above. But, in addition, the

interwar labor market differed from the present by the extent of low job

attachment. It was more common than today for workers to change

employers. The extreme case was that of "casual workers," who might

queue up at a different factory gate each morning to be taken on for only

a day or even a morning's work. It followed that workers without firm job

attachments might move in and out of the pool of unemployed while moving

between employers.

The average spell of unemployment was longer than the experience of

casual workers suggests, however. In the 1930s an unemployed person might

expect to remain out of work for 8 to 13 weeks. But different groups

remained in unemployment for very different spells of time. At least a

quarter of those entering unemployment could expect to remain there for

more than a year. This likelihood grew as the depression persisted:

long-term unemployment rose from less than 5 per cent of total

unemployment in 1929 to 22 per cent between 1933 and 1937. The danger of

long-term unemployment was greatest for older males. Older workers did

not face a greater risk of losing their jobs, but once in unemployment

they had dimmer prospects of getting out.

Thus, unemployment in interwar Britain encompassed a combination of

two very different phenomena. On the one hand were a large number of

persons repeatedly moving into and out of the pool of unemployment and

experiencing unemployment spells of short duration. The insurance system

was partly but not entirely responsible for this extraordinary amount of

"churning" in the labour market. On the other hand were a smaller number

of persons, mainly older men, who when unfortunate enough to enter

unemployment experienced very long spells without work. The insurance

system had little to do with their plight. Recent research shows

convincingly how misleading it is to lump the two groups together.

* * *

Why should some workers, once in unemployment, have found it

exceptionally difficult to escape? One possibility is that they differed

in their susceptibility to unemployment's debilitating effects. Older men

were thought to be particularly vulnerable to psychological impairment.

Long-time wage earners who took pride in their occupation found it

especially difficult to deal with being out of work. Household heads

despaired for having failed their families and coped with their feelings

by falling into a state of indifference and lassitude. While factors such

as these are difficult to document, recent historians like Harris [4] have

attached considerable importance to their effects.

It might seem that the impact of unemployment on physical well-being

would be comparatively easy to document. Yet contrary to the impressions

of laymen, throughout the interwar period Ministry of Health experts

adamantly denied that unemployment had adverse effects on health. They

were unconvinced by the observation that illness was concentrated where

unemployment was highest: areas with high unemployment were also ones with

long histories of poor health standards. They noted that neither the

crude death rate or the infant mortality rate rose significantly between

the 1920s and the 1930s. Many historians concur in this view: the work of

10

Jay Winter [7], for example, suggests that throughout the period

improvements in medical care supported a trend toward better health.

Many observers cannot help but feel that this view is too sanguine.

They will find support in Bernard Harris' [4] innovative analysis of the

problem. Harris draws on the physiological literature on nutrition and

stature, which analyzes how malnutrition can stunt the growth of children.

Using evidence gathered by the School Medical Officers on the average

height of schoolchildren between 1923 and 1938, Harris documents a

negative relationship between children's stature and local unemployment

for a number of middle-income towns (but no effect for towns where

unemployment was highest). He suggests that the level of unemployment

benefit was too low to enable the unemployed and their families to satisfy

all their physiological needs, but it was high enough to prevent a

deterioration in health in areas where the average level of nutrition was

already very low. Thus, earlier observers may have overlooked the

relationship between unemployment and physical well-being because

unemployment did not have its worst effects on the poorest groups.

In addition to providing new evidence on the effects of unemployment,

this new work has implications for its causes. For it is hard to

reconcile evidence that unemployment led to malnutrition and stunted

growth with the presumption that it was a voluntary phenomenon induced by

the operation of a generous dole. In addition, Harris' evidence that

rising unemployment was associated with deteriorating health for some

income groups but not others shows again the dangers of generalizing about

unemployment between the wars. Interwar unemployment had neither a single

cause nor a single set of effects. The research now underway promises to

shed new light on the entire range of experience.

11

REFERENCES

[1] Daniel K. Benjamin and Levis A. Kochin, "Searching for an Explanationfor Unemployment in Interwar Britain," Journal of Political Economy 87(1979), pp.441-478.

[2] Barry Eichengreen, "Unemployment in Interwar Britain: Dole orDoldrums?" Oxford Economic Papers 39, pp. 597-623.

[3] Barry Eichengreen, "Juvenile Unemployment in 20th Century Britain: TheEmergence of a Problem," Social Research 54 (1987), pp.273-302.

[4] Bernard Harris, "Unemployment, Insurance and Health in InterwarBritain," in Barry Eichengreen and T.J. Hatton (eds), InterwarUnemployment in International Perspective (Martinus Nijhoff, 1988),pp.149-184.

[5] Carol Heim, "Structural Transformation and the Demand for New Labor inAdvanced Economies: Interwar Britain," Journal of Economic History 44(1984), pp.585-595.

[6] Mark Thomas, "Labour Market Structure and the Nature of Unemploymentin Interwar Britain," in Eichengreen and Hatton, pp.97-148.

[7] Jay Winter, "Infant Mortality, Maternal Mortality and Public Health inBritain in the 1930s," Journal of European Economic History 8 (1979),pp.439-462.

* * * .

Barry Eichengreen is Professor of Economics at the University ofCalifornia Berkeley. In addition to his work on interwar unemployment, hehas published Sterling in Decline (with Sir Alec Cairncross) and The GoldStandard in Theory and History.

13